Hepatic encephalopathy (HE) is a very common neuropsychiatric complication of cirrhosis. It manifests as either a subclinical syndrome only detectable by psychometric tests or with overt signs of neuropsychiatric disturbance. The pathogenesis of this reversible syndrome is unknown. Consequently therapy for this problem has remained largely empirical and no new treatment has been introduced in more than two decades. Recently it has been discovered that most patients with overt clinical HE have elevated levels of endogenous benzodiazepine like activity (EBZ) in their blood. In subclinical HE however EBZ levels are not apparently increased. Thus the aim of the proposed studies are as follows. In cirrhotic patients (primarily from alcohol plus/minus chronic hepatitis C) we propose to compare our current techniques for extracting and measuring EBZ in HE patients to methods commonly employed in toxicology laboratories. Our hypothesis is that standard toxicology techniques cannot reliably detect EBZ. Based on the recent identification of substantial EBZ in stools from patients with HE we propose also to characterize and measure the amount of EBZ in stool. The hypothesis is that raised levels of EBZ in the patients arises from the gut. Using stool examination or tissue and large blood samples obtained postmortem it is planned to formally identify the compounds responsible for EBZ in HE patients using chromatography and mass spectrometry and to indicate their source of origin. Finally two sets of clinical studies will be done in selected cirrhotic patients with subclinical and overt HE respectively. It is proposed to establish if elevated levels of ammonia, free tryptophan or EBZ correlate with the degree of severity of HE in longitudinal studies while patients are being treated. "Arterialized" venous blood sampling will be done to obtain a profile of the levels of those compounds being presented to the brain. In addition interactions between ammonia, free fatty acids, free tryptophan, EBZ and protein binding in blood will be explored. The hypothesis being tested in these protocols are that ammonia and free tryptophan may play a role in subclinical HE and early overt HE, but EBZ plays a major role in severe HE.